yerfor / MimicTalk

MimicTalk: Mimicking a personalized and expressive 3D talking face in minutes; NeurIPS 2024; Official code
MIT License
430 stars 47 forks source link

Error in sys.excepthook #10

Open nitinmukesh opened 1 month ago

nitinmukesh commented 1 month ago

Not sure what is the issue

(mimictalk) C:\ai\MimicTalk>python inference/app_mimictalk.py
args.a2m_ckpt checkpoints/240112_icl_audio2secc_vox2_cmlr
args.head_ckpt
args.torso_ckpt checkpoints_mimictalk/German_20s/model_ckpt_steps_10000.ckpt
1
| WARN: checkpoints\240112_icl_audio2secc_vox2_cmlr\audio2secc_vae.yaml not exist.
| load 'model' from 'checkpoints/240112_icl_audio2secc_vox2_cmlr\model_ckpt_steps_1856000.ckpt', strict=True
| WARN: checkpoints_mimictalk\German_20s\secc_img2plane.yaml not exist.
Error in sys.excepthook:

Original exception was:
nitinmukesh commented 1 month ago

Even the command line is not working. Error handling is bad in code, no clear information what is wrong.

(mimictalk) C:\ai\MimicTalk>python inference/train_mimictalk_on_a_video.py
copy "checkpoints/mimictalk_orig/os_secc2plane_torso\config.yaml" "checkpoints_mimictalk/GER"
        1 file(s) copied.
| WARN: checkpoints\mimictalk_orig\os_secc2plane_torso\secc_img2plane.yaml not exist.
zhanghongyong123456 commented 4 weeks ago

Even the command line is not working. Error handling is bad in code, no clear information what is wrong.

(mimictalk) C:\ai\MimicTalk>python inference/train_mimictalk_on_a_video.py
copy "checkpoints/mimictalk_orig/os_secc2plane_torso\config.yaml" "checkpoints_mimictalk/GER"
        1 file(s) copied.
| WARN: checkpoints\mimictalk_orig\os_secc2plane_torso\secc_img2plane.yaml not exist.

you need config.yaml file ,the info warning ,Check the file or download it again

nitinmukesh commented 4 weeks ago

Config.yaml is already there

image

This file is missing secc_img2plane.yaml

https://drive.google.com/drive/folders/1KM0GubbN0B6l3CCAL31341TYaFmJCXAx

ultrafunkamsterdam commented 2 weeks ago

I've learned the hard way.

in short, what you need to do:

image

and off you go: image

nitinmukesh commented 2 weeks ago

@ultrafunkamsterdam

Appreciate your response. I gave up on this but with your help the gradio Ui is running.

However still getting error

| python data_gen/utils/process_video/extract_segment_imgs.py --ds_name=nerf --vid_dir=data/raw/videos/German_20s.mp4
todo videos number: 1
INFO: Created TensorFlow Lite XNNPACK delegate for CPU.
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
W0000 00:00:1730815048.379380    2164 inference_feedback_manager.cc:114] Feedback manager requires a model with a single signature inference. Disabling support for feedback tensors.
| Extracting Segmaps && Saving...
<class 'OSError'> [WinError 6] The handle is invalid
| Head images Extracted!
create rasterizer on device cuda:0
Training ERROR: 'NoneType' object is not subscriptable
ultrafunkamsterdam commented 2 weeks ago

@ultrafunkamsterdam

Appreciate your response. I gave up on this but with your help the gradio Ui is running.

However still getting error

| python data_gen/utils/process_video/extract_segment_imgs.py --ds_name=nerf --vid_dir=data/raw/videos/German_20s.mp4
todo videos number: 1
INFO: Created TensorFlow Lite XNNPACK delegate for CPU.
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
W0000 00:00:1730815048.379380    2164 inference_feedback_manager.cc:114] Feedback manager requires a model with a single signature inference. Disabling support for feedback tensors.
| Extracting Segmaps && Saving...
<class 'OSError'> [WinError 6] The handle is invalid
| Head images Extracted!
create rasterizer on device cuda:0
Training ERROR: 'NoneType' object is not subscriptable

i haven't tried extract_segment_imgs.py but only app_mimictalk.py . any way. if you have ipython installed, open it up, type %pdb
and on the next line %run data_gen/utils/process_video/extract_segment_imgs.py --ds_name=nerf --vid_dir=data/raw/videos/German_20s.mp4

it is more easy to locate what is wrong.